The present invention relates generally to diagnostic imaging and, more particularly, to a stacked detector having a logic control to select data usage of a pair of detectors.
Typically, in radiographic imaging systems, such as x-ray and computed tomography (CT), an x-ray source emits x-rays toward a subject or object, such as a patient or a piece of luggage. Hereinafter, the terms “subject” and “object” may be interchangeably used to describe anything capable of being imaged. The beam, after being attenuated by the subject, impinges upon an array of radiation detectors. The intensity of the attenuated beam radiation received at the detector array is typically dependent upon the attenuation of the x-rays. Each detector element of the detector array produces a separate electrical signal indicative of the attenuated beam received by each detector element. The electrical signals are transmitted to a data processing system for analysis which ultimately produces an image.
In other typical radiographic imaging systems, positron emission tomography (PET) or single photon emission computed tomography (SPECT) a radiation source within the imaged object emits x-rays which are intercepted by a photon counting, energy sensitive x-ray detector. A CT system can be paired with a PET or SPECT system to produce a fused system (CT/SPECT or CT/PET) providing images indicating both anatomical structure and physiologically significant (i.e. functional) information. Such combined systems include a source that emits x-rays toward a x-ray detector and separate SPECT or PET detector which measures x-rays emitted from radiation source within the object.
In some CT imaging systems, for example, the x-ray source and the detector array are rotated within a gantry and within an imaging plane around the subject. X-ray sources for such CT imaging systems typically include x-ray tubes, which emit the x-rays as a fan beam emanating from a focal point. X-ray detectors for such CT imaging systems typically are configured in an circular arc centered to the focal spot. In addition, such detectors include a collimator for collimating x-ray beams received at the detector with focus to the focal spot. In addition, such detectors include a scintillator for converting x-rays to light energy adjacent the collimator, and a photodiode for receiving the light energy from an adjacent scintillator and producing electrical signals therefrom. Typically, each scintillator of a scintillator array converts x-rays to light energy. Each photodiode detects the light energy and generates a corresponding electrical signal as a function of the light emitted by a corresponding photodiode. The outputs of the photodiodes are then transmitted to the data processing system for image reconstruction.
In some SPECT and PET systems, for example, the one or more flat detector arrays is rotated within a gantry and within an imaging plane and around the subject. X-ray radiation sources within the imaged object emit photons in random directions. A x-ray detector typically includes a collimator for collimating x-ray beams received at the detector with focus for parallel rays contained within the imaging plane and perpendicular to the detector plane. In addition, such detectors include a scintillator for converting x-rays to light energy adjacent the collimator, and a photomultiplier tube for receiving the light energy from an adjacent scintillator and producing electrical signals therefrom which are then transmitted to the data processing system for image reconstruction.
Conventional CT imaging systems utilize detectors that convert radiographic energy into current signals that are integrated over a time period, then measured and ultimately digitized. A drawback of such detectors however is their inability to provide data or feedback as to the number and/or energy of photons detected. That is, conventional CT detectors have a scintillator component and photodiode component wherein the scintillator component illuminates upon reception of radiographic energy and the photodiode detects illumination of the scintillator component and provides an electrical signal as a function of the intensity of illumination. While it is generally recognized that CT imaging would not be a viable diagnostic imaging tool without the advancements achieved with conventional CT detector design, a drawback of these detectors is their inability to provide energy discriminatory data or otherwise count the number and/or measure the energy of photons actually received by a given detector element or pixel. That is, the light emitted by the scintillator is a function of the number of x-rays impinged as well as the energy level of the x-rays. Under the charge integration operation mode, the photodiode is not capable of discriminating between the energy level or the photon count from the scintillation. For example, two scintillators may illuminate with equivalent intensity and, as such, provide equivalent output to their respective photodiodes. Yet, the number of x-rays received by each scintillator may be different as well as the x-rays intensity, but yield an equivalent light output.
A typical PET or SPECT system uses a photon counting, energy discriminating detector constructed from a scintillator and photomultiplier tube. Such detectors have large detector elements and as such are not readily adapted to CT applications requiring high resolution imaging to capture anatomical detail in the imaged object. Accordingly, recent detector developments have included the design of an energy discriminating, direct conversion detector that can provide photon counting and/or energy discriminating feedback with high spatial resolution. In this regard, the detector can be caused to operate in an x-ray counting mode, an energy measurement mode of each x-ray event, or both.
Energy discriminating, direct conversion detectors are capable of not only x-ray counting, but also providing a measurement of the energy level of each x-ray detected. Consequently, such a detector could potentially be used for SPECT or PET imaging. While a number of materials may be used in the construction of a direct conversion energy discriminating detector, semiconductors such as Cadmium Zinc Telluride (CZT), Cadmium Telluride (CdTe) and the like have been shown to be preferred materials.
Generally, in order to create an image using a direct conversion detector, either charge integration or charge pulse counting electronics are employed. For charge integration, the charge is integrated over the area of the pixel and the integrated charge is used to generate a digital signal indicative of total x-ray count and the energy level of each x-ray detected. For pulse counting, the amplitude of the current pulse is compared to a threshold level and stored in bin counters in order to record photon energy.
A drawback of direct conversion semiconductor detectors, however, is that these types of detectors cannot count at the x-ray photon flux rates typically encountered with conventional CT systems, e.g. at or above 5-100 million counts per sec per millimeter squared (Mcps). The very high x-ray photon flux rate, above 5-100 Mcps, causes pile-up and polarization which ultimately leads to detector saturation. That is, these detectors typically saturate at relatively low x-ray flux level thresholds. Above these thresholds, the detector response is not predictable or has degraded dose utilization. For SPECT and PET, imaging flux levels are below 5-100 Mcps and such saturation in a semiconductor detector for SPECT and PET is not a practical concern. However, for CT, saturation can occur at detector locations wherein small subject thickness is interposed between the detector and the radiographic energy source or x-ray tube. It has been shown that these saturated regions correspond to paths of low subject thickness near or outside the width of the subject projected onto the detector fan-arc. In many instances, the subject is more or less circular or elliptical in the effect on attenuation of the x-ray flux and subsequent incident intensity to the detector. In this case, the saturated regions represent two disjointed regions at extremes of the fan-arc. In other less typical, but not rare instances, saturation occurs at other locations and in more than two disjointed regions of the detector. In the case of an elliptical subject, the saturation at the edges of the fan-arc is reduced by the imposition of a bowtie filter between the subject and the x-ray source. Typically, the filter is constructed to match the shape of the subject in such a way as to equalize total attenuation, filter and subject, across the fan-arc. The flux incident to the detector is then relatively uniform across the fan-arc and does not result in saturation. What can be problematic, however, is that the bowtie filter may not be optimal given that a subject population is significantly less than uniform and not exactly elliptical in shape. In such cases, it is possible for one or more disjointed regions of saturation to occur or conversely to over-filter the x-ray flux and create regions of very low flux. Low x-ray flux in the projection will ultimately contribute to noise in the reconstructed image of the subject.
Detector saturation causes loss of imaging information and results in artifacts in x-ray projection and CT images. In addition, hysteresis and other non-linear effects occur at flux levels near detector saturation as well as flux levels over detector saturation. Direct conversion detectors are susceptible to a phenomenon called “polarization” where charge trapping inside the material changes the internal electric field, alters the detector count and energy response in an unpredictable way, and results in hysteresis where response is altered by previous exposure history. In particular, photon counting, direct conversion detectors, saturate due to the intrinsic charge collection time (i.e. dead time) associated with each x-ray photon event. Saturation will occur due to pulse pile-up when x-ray photon absorption rate for each pixel is on the order of the inverse of this charge collection time. The charge collection time is approximately proportional to thickness of the direct conversion layer for a fixed electric field and anode contact size; therefore, an increase in saturation rate is possible if the direct conversion layer is thinner. However, a sufficient thickness is required to stop almost all the x-rays. Incomplete collection of x-rays results in reduced image quality, i.e. a noisy image, and poor utilization of dose to the imaged object.
An additional factor in the charge collection time is the voltage applied across the layer thickness. A larger electric field (voltage/thickness) results in inverse proportionally smaller charge collection times and proportionally larger saturate rates. However, there is a reliability issue to routing of high voltage signals. Higher reliability can be obtained for lower voltages across smaller thicknesses of direct conversion layer. However, again, a sufficient thickness of the layer is required to sufficiently stop a majority of the x-rays.
Other types of detectors in addition to direct conversion detectors also saturate. A common example is the scintillator-photodiode arrangement connected to an integrating preamplifier. Charge created from each photon is routed to the preamplifier. As x-ray flux increases, the current to the preamplifier or total charge built up over an integration time period will increase. The readout electronics have a limiting current or charge capability before saturating the amplifier. Amplifier saturation is associated with non-linear response and the loss of signal charge. This again results in poor dose utilization and image artifacts.
Another detector construction is a scintillator over photodiode connected to photon counting readout electronics. Similar constructions utilize a scintillator over avalanche-photodiode or photo-multiplier tube. Saturation of the x-ray flux rate in these photon-counting cases is also related to a dead time for clearing the charge before arrival of the next x-ray photon.
For photon counting, direct conversion detectors, a practical solution to x-ray flux rate saturation in imaging systems using x-ray sources operating above the saturation point is not known. For these systems, a total thickness of the x-ray absorbing layer may be greater than, for instance, 1.0 mm. The higher the energy of the x-rays; the higher the required thickness to sufficiently stop a predominance of the x-ray flux. A typical target value is to stop 90-99% of the incident x-rays. For Cadmium Zinc Telluride (CZT) or Cadmium Telluride (CdTe), two possible direct conversion materials used for x-ray spectroscopy, the thickness for diagnostic radiology and CT imaging is approximately 3.0-5.0 mm in order to stop most of the x-rays generated from a source at 100-200 kVp. For CZT and CdTe, the saturation limit of roughly 5-100×106 x-rays/sec/mm2 is generally found for pixel size on the order of 1.0 mm and thicknesses of order 3.0-5.0 mm. This limit is directly related to the charge collection time for CZT. Higher flux rates are theoretically possible using smaller pixels. Each pixel has a size-independent count rate limit set by the charge collection time. The saturation flux rate is set by the count rate limit divided by the area of the pixel. Therefore, the saturation flux rate increases as the pixel size decreases. Smaller pixels are also desirable because they make available higher spatial resolution information which can result in high resolution images. However, small pixel size results in higher cost and there are more channels per unit area which need to be connected to readout electronics.
In addition, smaller pixels or detector elements have larger perimeter to area ratios resulting in more cross-talk. The perimeter is a region where charge is shared between two or more pixels (i.e. cross-talk). This sharing of charge results in incomplete energy information and/or a miscount of x-ray photons because the readout electronics are not configured to combine simultaneous signals in neighboring pixels. Very high flux rates are possible with thin, photon counting, direct conversion silicon layers with pixel size <0.1 mm, but there is not sufficient stopping power in these thin layers to stop the x-rays. For integrating detectors, the size of the detector pixel and design of the preamplifier are balanced to handle an x-ray flux rate expected during imaging. For CT, the flux rate capability of the detector with integrating electronics is generally of the order 109 photons/sec/mm2. For x-ray projection imagers operating with charge storage, integrating detectors, the flux rate capability is only of the same order. For photon counting detectors using scintillators and one of photodiodes/APDs/photomultipliers, the dead time of the x-ray conversion layer is very fast and the dead time is usually related to the bandwidth of the electronic readout, which can also be relatively high. The problem with these detectors is varied. In the case of a photodiode, the electronic gain is not sufficient to overcome the electronic noise. In the case of APDs, there is additional gain but with associated gain-instability noise, temperature sensitivity and reliability issues. In the case of photomultiplier tubes, these devices are too large and costly for high resolution detectors covering large areas.
Detector saturation can affect image quality by constraining the number of photons used to reconstruct the image and introducing image artifacts. A minimum image quality, therefore a minimum flux rate, is required to make use of the images. In this regard, when setting the configuration of the system such that sufficient flux is received at one area of the detector, then it is likely that another area of the detector will receive higher flux, and possibly, high enough to saturate the detector in this area. Higher flux in these other areas is not necessary for the image quality; however, the loss of data due to detector saturation may need to be addressed through correction algorithms in order to reduce image artifacts. For CT imaging, the reconstruction is not tolerant to missing or corrupted data. For example, if the center of the detector is illuminated with a minimum flux for image quality purposes, and if the illuminated object is compact, then detector cells at and beyond the periphery of the object's shadow can be saturated due to thin object thickness in these projected directions. The reconstruction of the data set with these uncorrected saturated values will cause severe artifacts in the image.
A number of imaging techniques have been developed to address saturation of any part of the detector. These techniques include maintenance of low x-ray flux across the width of a detector array, for example, by using low tube current or current that is modulated per view. However, this solution leads to increased scanned time. That is, there is a penalty that the acquisition time for the image is increased in proportion to the nominal flux needed to acquire a certain number of x-rays that meet image quality requirements.
With respect to combined CT and SPECT or CT and PET imaging, the availability of an energy discriminating detector with high flux rate capability provides the opportunity for a shared detector. The x-ray photon energies of SPECT are similar to those in CT, such that a semiconductor layer thickness can be designed to meet the requirements of both CT and SPECT. However, for PET, the photon energies are at 511 eV, about 5 times higher than that used for CT and SPECT.
In a typical imaging application, x-rays are absorbed in the direct conversion material which results in creation of an electrical charge in the direct conversion material. In order to create digital image information, the charge generated is collected on the segmented anodes typically using either charge integration or charge pulse counting electronics.
One drawback of direct conversion semiconductor detectors, however, is that x-rays absorbed in the direct conversion material near the gaps or perimeters of the anodes can result in a charge being generated therein that is shared by at least two neighboring pixel anodes. When using charge integration electronics, charge sharing can manifest itself as crosstalk between neighboring pixels, thus rendering the electronics susceptible to electronic noise amplification and spatially blurring of the image. When using pulse counting electronics, charge sharing can result in dividing the charge between at least two anodes, resulting in lost counts when the amplitude of the charge pulse collected in at least one of the anodes is below a discrimination threshold. Additionally, when pulse counting, high energy x-rays can result in loss of detection quantum efficiency (DQE) by the creation of two or more counts being collected in two or more neighboring anodes, thus mis-counting the events and binning for instance a single high energy event as two or more low-energy events.
Another drawback of direct conversion semiconductor detectors is that the response at the edge and corners of the direct conversion crystal may not be not reproducible. Such locations of a direct conversion crystal typically have charge trapping centers that cause changes in the internal electric field as the incident x-ray flux changes. The changing internal field can cause a poor detector response that can lead to image quality problems.
A number of techniques have been developed to address charge-sharing in direct conversion detectors. Energy discriminating detectors typically comprise a number of segmented anodes of typically 0.2-2.0 mm that define a pixelated structure onto which the direct conversion material is electrically attached. The anodes define the response area of the imaging pixels which segment the area of the detection plane. Smaller pixels are generally desirable because they make available higher spatial resolution information which can result in higher resolution images, and because the flux rate capability is generally improved with smaller pixels. However, smaller pixel size can result in higher cost because there are more channels per unit area which need to be connected to readout electronics. In addition, smaller pixels or detector elements have larger perimeter to area ratios resulting in a larger percentage of charge sharing regions per unit area of the detector.
Because the perimeter is the region where a charge may be shared between two or more pixels, incomplete energy information and/or a miscount of x-ray photons occurs because the readout electronics are not configured to combine near-simultaneous signals in neighboring pixels. Readout electronics could incorporate a time-coincidence circuit configured to identify events occurring within a defined time window that, once identified, prevents the detected event from receiving a bin count. However, such electronics can be costly and difficult to implement.
To solve the problem regarding the reproducibility of the response at the edge and corners of the direct conversion crystal, a guard ring is typically placed on the anode surface of the device or on side walls of the crystal walls. However, a guard ring does not prevent trapping of charge within the semiconductor and does not prevent a changing electric field from developing within the semiconductor.
It would therefore be desirable to design a direct conversion, energy discriminating CT detector that can produce images above x-ray photon flux rates that typically cause saturation within a direct conversion detector. It would be further desirable to design an x-ray management system that accommodates variations in x-ray flux across a CT detector assembly and compensates for over-ranging or saturating detectors, and precludes charge-sharing within a direct conversion detector. Such a detector and flux management system would allow the use of the same detector for both CT and SPECT imaging.